Metadata
Description
Using the current model structure used by the video metada panel as an example:
{
url: "",
highContrast: false,
signLanguage: false,
flashing: "unknown",
audio: "available",
keywords: [],
captions: []
}
This structure doesn't make much sense to integrators. A more easy-to-understand structure would apply to the metadata output such as:
{
url: "http://url",
metadata: {
accessModel: [], // An array of any of these values: "visual", "audio"
accessibilityHazard: [], // An array of one of these values: "flashing" or "noFlashing"
accessibilityFeature: [], // An array of any of these values: "highContrast", "signLanguage", "captions"
keywords: [], // An array of any of these values: "dialogue", "soundtrack", "sound effect"
},
captions: [{
src: "http://captionUrl",
language: ".."
},
....
]
}
So, a central model with this structure can be exposed to integrators and use model transformation to transform it to the models required by media panels.
Comments
-
Cindy Li commented
2014-06-02T12:31:59.808-0400 Submitted a pull request: https://github.com/fluid-project/metadata/pull/25
-
Justin Obara commented
2014-08-05T08:45:48.251-0400 Pull request ( https://github.com/fluid-project/metadata/pull/25 ) merged at 400487d4b7ccc1ec749e41c56f765f572029565b